GeometryForcing/algorithms/vae/common/distribution.py

from typing import List
import torch
import numpy as np


class DiagonalGaussianDistribution(object):
    def __init__(self, parameters, deterministic=False):
        self.parameters = parameters
        self.mean, self.logvar = torch.chunk(parameters, 2, dim=1)
        self.logvar = torch.clamp(self.logvar, -30.0, 20.0)
        self.deterministic = deterministic
        self.std = torch.exp(0.5 * self.logvar)
        self.var = torch.exp(self.logvar)
        if self.deterministic:
            self.var = self.std = torch.zeros_like(self.mean).to(
                device=self.parameters.device
            )

    def sample(self):
        x = self.mean + self.std * torch.randn(self.mean.shape).to(
            device=self.parameters.device
        )
        return x

    def kl(self, other=None):
        if self.deterministic:
            return torch.Tensor([0.0])
        else:
            dim = [1, 2, 3] if self.mean.dim() == 4 else [1, 3, 4]  # BCHW or BCTHW
            if other is None:
                return 0.5 * torch.sum(
                    torch.pow(self.mean, 2) + self.var - 1.0 - self.logvar,
                    dim=dim,
                )
            else:
                return 0.5 * torch.sum(
                    torch.pow(self.mean - other.mean, 2) / other.var
                    + self.var / other.var
                    - 1.0
                    - self.logvar
                    + other.logvar,
                    dim=dim,
                )

    def nll(self, sample, dims=[1, 2, 3]):
        if self.deterministic:
            return torch.Tensor([0.0])
        logtwopi = np.log(2.0 * np.pi)
        return 0.5 * torch.sum(
            logtwopi + self.logvar + torch.pow(sample - self.mean, 2) / self.var,
            dim=dims,
        )

    def mode(self):
        return self.mean

    @classmethod
    def cat(cls, distributions: List["DiagonalGaussianDistribution"], dim: int = 0):
        """
        Concatenates a list of DiagonalGaussianDistributions along the batch dimension.
        """
        parameters = torch.cat([dist.parameters for dist in distributions], dim=dim)
        all_deterministic = all([dist.deterministic for dist in distributions])
        any_deterministic = any([dist.deterministic for dist in distributions])
        assert (
            all_deterministic == any_deterministic
        ), "`deterministic` must be the same for all distributions when concatenating."
        return cls(parameters, deterministic=all_deterministic)